^{page 385 note 1} Ellis, Robinson, The Fables of Avianus (Oxford, 1887), pp. xi–xxxix,Google ScholarHervieux, L., Les Fabulistes latins iii (Paris, 1894), pp. 1–48:Google Scholar cf. also Unrein, O., De Aviani aetate (Diss., Jena, 1885), passim.Google Scholar

^{page 385 note 2} Op. cit., p. xxi: for Theodosius’ antipagan legislation and its repercussions in the West see most recently Bloch, H., in The Conflict between Paganism and Christianity in the Fourth Century, ed. Momigliano, A. (Oxford, 1963), pp. 198 f.Google Scholar

^{page 385 note 3} For the history of the fable in the Graeco-Roman world see the preface to Perry's, B. E. new edition of Babrius and Phaedrus (Loeb, 1965), pp. xi–xcvi.Google Scholar

^{page 385 note 4} J. Straub has recently much exaggerated the extent to which pagans were prevented from expressing their views at this period (Heidnische Geschichtsapologetik in der christlichen Spätantike, Bonn, 1963,Google Scholarpassim: see my objections in JRS lv [1965], 241–2).Google Scholar

^{page 386 note 1} Cf.Fargues, P., Claudien (Paris, 1933), p. 21,Google Scholar and Iraldi, M., Atti della Accademia delle Scienze di Torino xcviii (1963-1964), 24. I can see no likelihood at all in Ellis's claim (op. cit., p. xix) that Ausonius to. 41 is an echo of Avianus 26. 11–12 (cf. Appendix B), nor that Symmachus at Ep. 1. 101 is alluding to Avianus 29. 21–22. Similarly with Prudentius, Perist. to. 1104 f. and Avianus 39 (Ellis p. xx n. t). On the other hand I attach more importance than Ellis to the parallels between Macrobius and Avianus: see below, p. 388 n. 6.Google Scholar

^{page 386 note 2} The addition of imperatorem after Theodosium in the tituli to some manuscripts must be dismissed as a scribe's guess based on the coincidence of name. It is out of the question that an emperor could be addressed in such a familiar way at this period. Moreover, it is clear from the tone of the preface as a whole that the dedicatee is a literary man.

^{page 386 note 3} ‘The Date and Identity of Macrobius’, JRS lvi (1966), 25–27.Google Scholar

^{page 386 note 4} Keil, , GL v. 599. Cf. my note in BICS xiv (1967).Google Scholar

^{page 387 note 1} Courcelle, P., Les Lettres grecques en occident de Macrobe d Cassiodore² (Paris 1948),Google Scholar ch. i: I am not convinced by E. Türk's recent attempt to prove that Macrobius obtained much of his early material, both Greek and Latin, at first hand (Latomus xxiv [1965], 381 f.)Google Scholar If Macrobius got most of his knowledge of early Latin poetry at third or fourth hand (as recently demonstrated by Jocelyn, H. D., C.Q.N.S. xiv [1964], 280–95, and N.S. XV [1965], 126–44), then a fortiori he is not likely to have had much first-hand knowledge of classical Greek literature.CrossRefGoogle Scholar

^{page 387 note 2} Courcelle, op. cit., part i, passim.

^{page 387 note 3} Montana, M. F., ‘Simmaco e la cultura greca’, Rend. 1st. Lombardo, cl. di. lett., cxv (1963),Google Scholar 297 f., Kroll, W., ‘De Q. Aur. Symm. studiis graec. et latinis’, Bresl. Philol. Abh. vi. 2 (1891), 6–16,Google Scholar and cf. also JRS liv (1964), 17.Google Scholar

^{page 387 note 4} It is of course out of the question that Macrobius should not have been a Roman citizen in the technical sense 200 years after the Constitutio Antoniniana.

^{page 387 note 5} Sat. I. I. 11–12.

^{page 387 note 6} ?Cf. C.Q. N.S. xiv (1964), 324–5.Google Scholar

^{page 387 note 7} Stahl, W. H., Macrobius' Commentary on The Dream of Scipio (N.Y., 1952), p. 5.Google Scholar

^{page 387 note 8} See the authorities cited by Stahl, op. cit., p. 4.

^{page 388 note 1} Türk, E., REL xli (1963), 348:Google Scholar see JRS lvi (1966) 34f.Google Scholar

Stahl's claim, op. cit. (p. 387 n. 7), p. 9, that Macrobius ‘certainly could have been a Christian’ cannot be entertained. It is impossible to believe that any Christian could have said that Vettius Praetextatus, notorious as the leader of the pagan reaction of his day, was ‘unum arcanae deorum naturae conscium, qui Bolus divina et adsequi animo et eloqui posset ingenio’ (Sat. I. 24. I): cf. also 1. 7. 17, ‘sacrorum … omnium Vettius unite conscius’ (similar descriptions at I. 17. I, and I). To St. Jerome he was ’homo sacrilegus et idolorum cultor‘ (c. Joh. Hieros. 8). Cf. Addendum, p. 399.

^{page 388 note 2} JRS lvi (1966) 25–38,Google Scholarpassim, developing a suggestion of Mazzarino, S., Rend. 1st. Lombardo, 1938, p. 257.Google Scholar

^{page 388 note 3} Brown, P. R. L., ‘Aspects of the Christianisation of the Roman Aristocracy’, JRS li (1961), 1f.Google Scholar

^{page 388 note 4} Comm. I. 2. 7 f.: cf. Stahl, op. cit., p. 84 n. 2.

^{page 388 note 5} There is no such defensive attitude in Babrius’ preface, and such as there is in the prefaces to the various books of Phaedrus, concerns the literary, not moral, character of his fables.

^{page 389 note 1} Ellis is too kind to Avianus in suggesting that he had in mind Phaedo 60–61, where Socrates relates how he had once turned some Aesopic fables into verse. If Avianus preferred to base his fables on what he admits was a poor Latin version instead of directly on the clear, easy Greek of Babrius (see Appendix B), then it is unlikely that he had either the ability or the inclination to read Plato. I suggest that he was misled by references to Socratici libri (Cicero, de orat. 3. 67, Horace C. 3. 21. 9) and Socraticae chartae (Horace, A.P. 310) into supposing that Socrates himself had written the books which contained his teachings. St. Jerome madea not dissimilar error at about the same time, when he pretended that he derived the knowledge of Pythagorean teachings he in fact owed to Cicero and Seneca from the original works of Pythagoras (Ep. 84. 6. 2). His rival Rufinus was quick to point out that there were no original writings of Pythagoras: Jerome could only reply, rather lamely, ‘de dogmatibus eorum, non de libris locutus sum, quae potui in Cicerone, Bruto ac Seneca discere’ (Apol. adv. Ruf. 3. 39: Cf.Hagendahl, H., Latin Fathers and the Classics [1958], pp. 176–7).Google Scholar

^{page 389 note 2} Wissowa, G., De Macrobii Sat. fontibus, Diss. Breslau, 1880, p. 12.Google Scholar

^{page 389 note 3} Jan, So, in his edition, vol. i (1848), p. xiii, Bornecque, note ad loc.Google Scholar

^{page 390 note 1} Not 115: as Browning, R. (C.R. N.S. x [1960], 42)Google Scholar and Dilke, O. A. W. (Latomus xviii [1959], 466) pointed out independently, Guaglianone's b² and Ln are one and the same manuscript, BM Reg. 15A.xxxi.Google Scholar

^{page 391 note 1} Handschriften antiker Autoren in mittelalterlichen Bibliothekskatalogen (Zentralblatt fur Bibliothekswesen, Beiheft 67), Leipzig, 1935, PP. 234–7.Google Scholar

^{page 391 note 2} It might be added that John of Salisbury employed the form Avienus: Opera, ed. Giles, iv. 189 (Cf.Manitius, , Philologus li [1892], 535).Google Scholar

^{page 392 note 1} Hervieux, , Fabulistes Latins, iii. 23.Google Scholar

^{page 392 note 2} There is another such confusion in the matter of a proper name in the manuscripts of the Sat. The man whom most manuscripts most of the time call Furius Albinus can only be Rufius Albinus, prefect of Rome in 389–91 (Cf.Chastagnol, A., Les Fastes de la prefecture de Rome au Bas-Empire [Paris, 1962], pp. 233–6). Rufius is guaranteed by several contemporary inscriptions (CIL vi. 3791a, 3791b, 36959). This fact seems to have escaped the notice of J. Willis, who in his new Teubner edition prints Furius throughout, even where (as at 1. 2. 16, the first occurrence of the name) all manuscripts give Rufius. Willis also gives Macrobius’ own name incorrectly as Ambrosius Theodosius Macrobius instead of Macrobius Ambrosius Theodosius—and repeats the error at the top of every other page throughout his edition: see JRS, 1966, p. 26.Google Scholar

^{page 392 note 3} Since it has passed through every edition of a widely read standard work, it may be useful to draw attention to the curious blunder of T. R. Glover (Virgil, p. 41 of the fourth edition), who writes of Avianius ‘the father of Symmachus’. Symmachus’ father was indeed called Avianius (in full L. Aurelius Avianius Symmachus), but Macrobius’ interlocutor was certainly not (see below, p. 403). Moreover, Macrobius repeatedly refers to the extreme youth of his Avienus, which would rule out the possibility that he was meant to be the father of Symmachus, in his mid forties at the dramatic date of the Sat., and in any case Avianius Symmachus died in 376, 8 years before the dramatic date.

^{page 392 note 4} The relevant part of the inscription runs as follows: ‘Abienii. eximiae integritatis viro ac mir(a)e bonitatis exemplo, Postumio Rufio Festo …’ It was standard practice for the signum to be placed at the head of such dedications in the genitive (Cf.Cagnat, R., Lours d' épig. latine⁴ [1914], pp. 53 f.). There can be no doubt that this is the poet. He is known to have held two proconsulates (ILS 2944); the Greek inscription cited above attests one, this new inscription the other. It is interesting to discover that he also bore the name Postumius: Rufius Postumius Festus, consul in 439, is clearly a direct descendant. I am grateful to Mr. J. Martindale for drawing my attention to the existence of this inscription.Google Scholar

^{page 393 note 1} Cf.de-Woestijne, P. van, La Descr. orbis terrae d' Avienus (1961), p. 10.Google Scholar

^{page 393 note 2} As I hope to have shown in JRS, 1966, P. 29.

^{page 393 note 3} ‘Zur Bestimmung der Zeit des Servius’, Philologus lxxi (1912), pp. 518 f.Google Scholar

^{page 393 note 4} For the date of Avien(i)us see a forthcoming article by Matthews, J. F. in Historia xvi (1967).Google ScholarGarroni, A., Bull. d. com. arch. com. xliii (1916), 123 f.,Google Scholar and Groag, E., Reichsbeamten von Achaia in spätrömischer Zeit (Diss. Pann. I. 14, 1946), p. 49, are no longer satisfactory.Google Scholar

^{page 394 note 1} Cf. CIL xi. 1. 2997, 2998, xv. 2. 7525: and Matthews, op. cit.

^{page 394 note 2} R.-E. ii. 2387, and cf. Groag, loc. cit. (P. 393 n. 4)

^{page 394 note 3} Venustus: cf. Sat. 1. 5. 53.

^{page 394 note 4} Cf. ILS 2944, ‘numeroque frequenti / natorum exultans’.

^{page 394 note 5} The son who erected his epitaph was called Placidus (ILS 2944): at least one other will presumably have been a Festus.

^{page 394 note 6} Cf. Ellis, p. xiv, ‘no remnant of Roman literature is more informed with the diction of Vergil than the Fables’.

^{page 394 note 7} This can hardly mean that he actually translated the whole of Livy into iambics in the way that Marianus of Eleutheropolis rather later in the century translated all the Hellenistic hexameter poets into iambics (Cf.Historia xiv [1965], 482). Surely all that he did was to put into iambics some of the stories from Livy, like the ‘Vergilii fabulas’ mentioned on Aen. to. 272. Versified Eutropius rather than versified Livy.Google Scholar

^{page 394 note 8} In all fairness it must be admitted that there are two objections to this identification: (a) it would entail, given my date for the publication of the Fables, a later date for the publication of Servius’ commentary than hitherto entertained. But the date of Servius is quite uncertain, and I have suggested elsewhere on other grounds that a later date would be preferable (JRS, 1966). (b) On Aen. 10. 272 Servius goes on to refer to his Avienus on the subject of comets: as Dr. S. Weinstock pointed out to me, this might seem to suit the author of the Aratea rather than the fabulist. But then the references to fabulae suit the fabulist better.

^{page 395 note 1} But not, as Chastagnol suggests (Fastes, cited at p. 392 n. 2 above, p. 235) the Albinus to whom Servius dedicated his Centimeter. See JRS, 1966, p. 30.

^{page 395 note 2} Cf.Martin, J., Symposion: die Geschichte einer literarischen Form (1931), pp. 79 f. Willis prints ‘Disarius’ throughout, despite the fact that almost all manuscripts offer ‘Dysarius’. It is only a small point, but the manuscripts of Symmachus’ Epp., while offering the name in a slightly corrupted form (Dysanus, Lysanus) at Ep. 3. 37, nevertheless point to the spelling with a y (though at Ep. g. 44 they give Disarius). Moreover, it is clear from Macrobius that the man was a Greek. Seeck, on the advice of Mommsen, printed Dusarius in both places in Symmachus.Google Scholar

^{page 395 note 3} Martin, op. cit., 71 f.

^{page 395 note 4} The case of Horus shows that even the minor characters of the Sat. are historical. He is also mentioned by Symmachus (Ep. 2. 39) and Libanius (Epp. 1278 and 1279 Forster). Libanius confirms that he was an Egyptian. Macrobius says that he was a boxer before he became a philosopher— an unlikely detail confirmed by Libanius, writing in 364, 20 years before the dramatic date of the Sat., who mentions that he won at the Olympic games of Antioch that year. Symmachus confirms that in later years he was a philosopher in Rome (the letter in question cannot be dated precisely, but must antedate 394).

^{page 396 note 1} Martin, op. cit., pp. 64 f.

^{page 396 note 2} That is, shortly before the date c. 431 or soon after, that I proposed for the Sat. in JRS, 1966. Professor Robert Browning and Mr. J. F. Matthews kindly read and commented on an earlier version of this paper, which was read in substantially this form to the Oxford Philological Society on 4 February 1966.

^{page 397 note 1} It seems legitimate to accent the word as in Greek this is the regular practice of Ammianus Marcellinus with latinized Greek proper names.

^{page 397 note 2} It would be unwise to tamper with the word-order to restore a clausula (e.g. ‘ridere animália fecimus’ [t]): the exception may be explained by the desire for chiasmus(‘certare volucres, animalia ridere’) overcoming that for a normal clausula. Such exceptions were expressly sanctioned by Cicero:‘quad numerosum in oratione dicitur non semper numero fiat, sed nonnumquam aut concinnitate aut construction verborum’ (Orator 202).